simple_app

Prerequisites

• Clone this repo and navigate to this directory.
• You need a python virtual environment. For this demo, Python 3.13.1 was used, but any Python environment 3.11+ should work.
• You will need Vespa CLI that you can deploy on MacOS with brew install vespa-cli
• Libraries dependencies can be installed with: pip install -R requirements.txt
• Sign-up with Tavily and Get an API key.
• Spin-up a Vespa Cloud Trial account:
  • Login to the account you just created and create a tenant at console.vespa-cloud.com.
  • Save the tenant name.
• A Valid OpenAI API key. Note that you have the option to use any other LLM, which may support Langgraph tools binding.
• Uncompress the data file: zstd -d data/vespa_feed-96k.json.zst

Deploy the Vespa Application

• Navigate to the app subdirectory
• Choose a name for your app. For example ecommercebot
• Follow instructions in the Getting Started document. Please note the following as you go through the documentation:
  • Skip Step 5. as you have already cloned the app directory.
  • You will need your tenant name you created previously.
  • When adding the public certificates with vespa auth cert, it will give you the absolute path of the certificate and the private key. Please note them down.
  • Return to the parent directory. To feed the application, run:

  cd ..
  vespa feed data/vespa_feed-96k.json
  

• You can test the following query from the Vespa CLI:

  vespa query "select id, category, title, price  from sources * where default contains 'screwdriver'"
  

• You will need the URL for the MTLS end-point for your Vespa application. Run the following command:

vespa status

This should return you an output like:

Container container at https://xxxxx.yyyyy.z.vespa-app.cloud/ is ready

Note down the URL for the MTLS end-point.

Configure and Launch your Streamlit Application

A template for secrets.toml file to store streamlit secrets has been provided. Please create a subdirectory .streamlit and copy the template there.

Update all the fields with all the information collected previously including the certicate locations and save the file as secrets.toml

Launch your streamlit application:

streamlit run streamlit_vespa_app.py

Test your Application

You can ask questions like:

What is the weather in Toronto ?

Followed by:

I need a screwdriver

Followed by:

How reliable is the TECKMAN brand ?

Followed by:

Which one would you recommend to repair a watch?

You will notice that the agentic application will leverage either a web search or query the e-commerce dataset stored on Vespa to address your queries seamlessly producing an engaging conversational experience.